Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Marker gene identification and functional characterization of dendritic cells (DCs)-like in turbot (Scophthalmus maximus).

Fish & shellfish immunology·2026
Same author

Extracellular β-glucan induces trained immunity in macrophages and is associated with anti-microbial defense in teleosts.

Fish & shellfish immunology·2026
Same author

PCLPred: identifying plant chloride transport-related proteins using reduced amino acid alphabets and N-peptide composition.

Amino acids·2026
Same author

Regional variation and prediction model of carbon emissions in the highway construction stage.

Scientific reports·2026
Same author

Palmitoylation in Renal Physiology and Pathology.

Biomolecules·2026
Same author

Transcriptomic analysis of a compatible tobacco-herbivore interaction and the role of jasmonoyl-L-isoleucine hydrolase 1 in response to growth/defense trade-off.

BMC plant biology·2025

相关实验视频

Updated: Jul 12, 2025

Author Spotlight: Advancing Large-Scale Neural Dynamics Through HD-MEA Technology
09:44

Author Spotlight: Advancing Large-Scale Neural Dynamics Through HD-MEA Technology

Published on: March 8, 2024

4.8K

多波长衍射神经网络使用权重方法.

Jianan Feng, Hang Chen, Dahai Yang

    Optics express
    |October 20, 2023
    PubMed
    概括

    研究人员开发了一种多波长衍射深度神经网络 (MW-D2NN),用于更快,更高效的AI任务. 这种新的衍射深度神经网络可以在各种光线条件下对图像进行分类,显示出先进机器视觉的前景.

    科学领域:

    • 光学和光子学 在光学和光子学.
    • 人工智能的人工智能
    • 计算科学 计算科学

    背景情况:

    • 分散深度神经网络 (D2NN) 为计算任务提供高速,低功耗和可扩展性.
    • 传统的D2NN通常是为单色光源设计的.
    • 对于能够在多波长照明下工作的DNN的需求正在增长.

    研究的目的:

    • 为多波长衍射深度神经网络 (MW-D2NN) 提出和展示一个框架.
    • 为了使D2NNN能够同时使用多个光波长进行计算.
    • 适应D2NN用于需要宽带或多色光源的应用.

    主要方法:

    • 开发了一种新的MW-D2NN框架,利用权重系数的方法.
    • 实施了一种训练策略,其中每个波长都被赋予特定的权重.
    • 基于不同波长的输出平面构建了一个波长加权损失函数.

    主要成果:

    • 受过训练的MW-D2NN成功地将手写数字图像分类在多波长事件光束下.
    • 一个三层的MW-D2NN实现了模拟分类准确率为83.3%.
    • 一个1层的MW-D2NN在RGB波长下展示了模拟和实验分类准确率分别为71.4%和67.5%.

    更多相关视频

    Automated Multimodal Stimulation and Simultaneous Neuronal Recording from Multiple Small Organisms
    08:28

    Automated Multimodal Stimulation and Simultaneous Neuronal Recording from Multiple Small Organisms

    Published on: March 3, 2023

    1.1K
    Revealing Neural Circuit Topography in Multi-Color
    09:11

    Revealing Neural Circuit Topography in Multi-Color

    Published on: November 14, 2011

    15.0K

    相关实验视频

    Last Updated: Jul 12, 2025

    Author Spotlight: Advancing Large-Scale Neural Dynamics Through HD-MEA Technology
    09:44

    Author Spotlight: Advancing Large-Scale Neural Dynamics Through HD-MEA Technology

    Published on: March 8, 2024

    4.8K
    Automated Multimodal Stimulation and Simultaneous Neuronal Recording from Multiple Small Organisms
    08:28

    Automated Multimodal Stimulation and Simultaneous Neuronal Recording from Multiple Small Organisms

    Published on: March 3, 2023

    1.1K
    Revealing Neural Circuit Topography in Multi-Color
    09:11

    Revealing Neural Circuit Topography in Multi-Color

    Published on: November 14, 2011

    15.0K

    结论:

    • 拟议的MW-D2NN框架有效地将衍射深度神经网络的能力扩展到多波长操作.
    • MW-D2NN显示了集成到在多波长和不连贯照明下运行的智能机器视觉系统的潜力.
    • 这项研究为更通用和更强大的衍射计算应用铺平了道路.